Whole saliva is gaining more and more attention as a diagnostic tool to study disease-specific changes in human subjects. Prior to the actual disease-related analyses, it is important to understand the influence of various demographic variables and coupled phenotypes on salivary protein signatures. In a cross-sectional approach, we analyzed the influence of age, sex, body mass index (BMI), smoking, and education on salivary protein signatures in whole saliva samples of 187 individuals. Subjects were randomly selected from the population-based Study of Health in Pomerania (SHIP-Trend). Stimulated whole saliva was collected, and proteins were precipitated and proteolytically digested. Samples were analyzed by label-free tandem mass spectrometry. Of the 602 human proteins identified in at least 40% of the saliva samples, we used 304 proteins, which could be identified with at least two unique peptides, for statistical analyses. Univariate and multivariate linear models were used to reveal associations with the phenotypes. The largest number of proteins was associated with smoking status. Moreover, age had a distinct influence on the salivary protein composition. The study discloses the influence of common phenotypes on the salivary protein pattern of human subjects. These results should be considered when studying disease-related proteome signatures in saliva.
Keywords: LC−MS/MS; label free quantitation; population-based study; protein signatures; whole saliva proteomics.